Influence of Bank Afforestation and Snag
Angle-of-fall on Riparian Large Woody
Debris Recruitment1
2
3
Don C. Bragg and Jeffrey L. Kershner
Abstract
A riparian large woody debris (LWD) recruitment simulator (Coarse Woody Debris [CWD])
was used to test the impact of bank afforestation and snag fall direction on delivery trends.
Combining all cumulative LWD recruitment across bank afforestation levels averaged 77.1
cubic meters per 100 meter reach (both banks forested) compared to 49.3 cubic meters per
100 meter reach (one side timbered). Both bank afforestation and snag fall patterns generated
significant differences in riparian LWD delivery, but there was no noticeable interaction.
Scenarios with only one bank forested delivered 15 to 50 percent less LWD than their two
bank counterparts. Snag fall patterns also produced statistically different LWD recruitment,
with some registering only 35 to 52 percent of the most productive fall patterns. These results
suggest testing the assumptions of random snag fall from two forested banks before modeling
riparian LWD recruitment.
Introduction
Large woody debris (LWD) recruitment is critical to healthy riparian ecosystems
(Bisson and others 1987, Dolloff 1994, Kershner 1997), making its recovery a
primary goal of streamside management (Berg 1995, Kershner 1997). Surprisingly
little work has been attempted on long-term recruitment dynamics, as research has
concentrated on the quantification of riparian LWD and its ecological role. However,
a growing interest in computer modeling of riparian LWD recruitment has prompted
the development of some simulators in recent years (e.g., Bragg and others 2000,
Murphy and Koski 1989, Rainville and others 1985, Van Sickle and Gregory 1990).
While creating a new riparian LWD recruitment model, we became concerned
about some traditional assumptions. For instance, most efforts have presumed that
both banks are equally forested. Although this may be true in mesic regions, some
semi-arid areas have limited forest on some banks. Modelers have also assumed
random snag fall without testing this premise. Random tree fall patterns can occur
when failure is not influenced by either disturbances or geomorphology (Maser and
Trappe 1984, Robison and Beschta 1990, Van Sickle and Gregory 1990). Although
most riparian LWD simulations have applied this pattern, other distinct
1
An abbreviated version of this paper was presented at the Symposium on the Ecology and Management
of Dead Wood in Western Forests, November 2-4, 1999, Reno, Nevada.
2
Graduate Research Assistant, Department of Forest Resources and Ecology Center, Utah State
University, Logan, UT 84322. Current address: Research Forester, Southern Research Station, USDA
Forest Service, P.O. Box 3516 UAM, Monticello, AR 71656 (e-mail: dbragg@fs.fed.us)
3
Fish Ecologist, USDA Forest Service, Fish Ecology Unit and Department of Fisheries and Wildlife,
Utah State University, Logan, UT 84322-5210 (e-mail: kershner@cc.usu.edu)
USDA Forest Service Gen. Tech. Rep. PSW-GTR-181. 2002.
65
Riparian Large Woody Debris Recruitment—Bragg and Kershner
configurations are possible (Alexander and Buell 1955, Bragg and others 2000,
Grizzel and Wolff 1998, Schmid and others 1985, Veblen 1986). Another biasing
factor, tree lean, plays only a limited role in riparian LWD delivery (Hairston-Strang
and Adams 1998). We decided to test the influence of different bank afforestation
and angle-of-snag-fall patterns on riparian LWD delivery to the stream using
computer simulation, which we hoped would improve the long-term prediction of
riparian LWD recruitment.
Methods
Project Design and Assumptions
The riparian LWD recruitment simulator Coarse Woody Debris (CWD, version
1.4) was used to predict bankfull channel delivery. CWD is a post-processor to the
Forest Vegetation Simulator (FVS) (Wykoff and others 1982). We will only briefly
describe the most salient features of the models’ interplay (Bragg and others 2000).
FVS establishes, grows, and kills all simulated trees, while CWD drives LWD
formation and channel recruitment. CWD takes dead trees, places them within the
riparian zone, selects an angle-of-fall from a predetermined distribution, fells and
breaks the snag, and assigns which pieces are recruited to the channel. Both
disturbance-related and self-thinning mortality can be emulated (Bragg 2000). CWD
randomly assigns tree locations in relation to the channel. Because the angle-of-fall
pattern set by the user is fixed for the whole riparian forest, biased fall directions
should be carefully designed to ensure consistency with local conditions. Depending
on the need, CWD allocated the trees to one or two banks. This effort assumed LWD
was greater than 10 centimeters in diameter and more than 1 meter long and was
“recruited” to the bankfull channel if it extended at least 1 meter into this zone.
Modeled Stream Description
Dry Lake Creek, a second order stream about 60 km northeast of Jackson,
Wyoming, on the Bridger-Teton National Forest, was exclusively used for these
simulations. Along the sampled reach, Dry Lake Creek had a mean bankfull width of
5.5 meters, an average gradient of 3.5 percent, a mean elevation of 2,565 meters, and
drained an upstream basin of 1,033 hectares (Bragg and others 2000). Riparian LWD
volumes within this reach of Dry Lake Creek averaged 8.6 cubic meters per 100
meter (Bragg and others 2000). The predominantly Engelmann spruce (Picea
engelmannii Parry ex Engelm.) and subalpine fir (Abies lasiocarpa [Hook.] Nutt.)
forests along Dry Lake Creek averaged approximately 33 square meters per hectare
of live basal area. Both banks along this particular reach were wooded, but for
demonstration purposes, half of the simulations considered only one side was
forested. A bankfull width of 5.5 meters and a streamside forest depth of 38 meters
for each bank were also assumed. All simulations covered 300 years.
Statistical Design and Analysis
In this study, we tested five different snag fall patterns: (1) random (RND); (2)
the tri-modal CWD default (DEF); (3) snags falling primarily towards the channel
(TWRD); (4) a fall pattern quartering towards the channel (QRT); and (5) snag
66
USDA Forest Service Gen. Tech. Rep. PSW-GTR-181. 2002.
Riparian Large Woody Debris Recruitment—Bragg and Kershner
failure largely paralleling the channel (PRL) under two bank afforestation conditions
(one [O] or both [B] forested) (table 1).
Table 1—Treatment codes, descriptions, and predicted riparian LWD recruitment by bank
afforestation and snag fall pattern.
Treatment
code
Description
Cumulative
volume
Standard
deviation
---- m3 per 100 m reach ----
2 banks forested
BRND
BDEF
BTWRD
BQRT
BPRL
Pooled
Random pattern
Tri-modal (CWD default) pattern
Fall direction towards the channel
Fall direction quartering towards channel
Fall direction parallel to the channel
Average of all 2 bank treatments
79.7
78.4
90.7
77.6
59.0
77.1
3.74
5.18
6.87
5.44
2.85
11.41
1 bank forested
ORND
ODEF
OTWRD
OQRT
OPRL
Pooled
Random pattern
Tri-modal (CWD default) pattern
Fall direction towards the channel
Fall direction quartering towards channel
Fall direction parallel to the channel
Average of all 1 bank treatments
39.9
38.4
76.6
64.0
27.5
49.3
3.80
2.60
2.00
4.59
2.16
20.27
Because of the stochasticity in some CWD subroutines, 10 replicates were run
for each snag fall pattern. Total LWD recruitment (in cubic meters per 100 meter
reach) over the simulation period was compared to determine the cumulative
significance of bank afforestation and snag fall patterns. Because of untransformable
heterogeneity of variance and non-normal data distributions, the nonparametric twofactor extension of the Kruskal-Wallis (K-W) analysis of variance test and a multiple
comparison using rank scores were used to identify treatment effects (Zar 1984).
Results
Both bank afforestation and snag fall direction significantly (P < 0.05) affected
cumulative LWD recruitment to Dry Lake Creek, but there was no significant
interaction between the two (table 1). Recruitment was always lower from streams
with one forested bank: when averaged across snag fall patterns, having both banks
forested delivered 77.1 cubic meters per 100 meter reach (standard deviation [SD] =
11.41), while one forested bank treatments averaged 49.3 cubic meters per 100 meter
reach (SD = 20.27).
With only one bank forested, random (ORND), default (ODEF), and OPRL snag
failure patterns yielded almost 50 percent less LWD recruitment than the same
patterns (BRND, BDEF, and BPRL) when both banks were forested. Of these
treatments, only the BPRL versus OPRL comparison proved statistically
insignificant. OTWRD and OQRT declined only 15 to 20 percent from BTWRD and
BQRT. However, consistently lower LWD delivery from only one forested bank
resulted in the rejection of the null hypothesis of no effect of bank afforestation.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-181. 2002.
67
Riparian Large Woody Debris Recruitment—Bragg and Kershner
The tree fall pattern predominantly towards the channel (TWRD) produced the
most LWD recruitment (regardless of bank afforestation), while the patterns
paralleling the channel (PRL) yielded the least (table 1). Even though BTWRD did
not noticeably differ from BQRT, BDEF, and BRND when both banks were forested,
it was significantly greater than BPRL. BTWRD (P < 0.05) and BRND (P < 0.10)
also contributed more LWD to the channel than BPRL. OTWRD yielded more LWD
than OPRL, ODEF, and ORND (35 percent, 50 percent, and 52 percent of OTWRD’s
cumulative LWD total, respectively), while OQRT provided more (P < 0.10) than
OPRL. The ORND, ODEF, and OPRL treatments did not differ statistically.
Discussion
Although these bank afforestation and snag fall patterns are greatly simplified,
their influence on LWD recruitment is statistically and ecologically meaningful.
Random or tri-modal (i.e., CWD default) patterns produced intermediate levels of
recruitment, while fall patterns biased strongly in particular directions resulted in
either greater or lesser delivery, depending on bank afforestation and the predominant
snag failure direction.
With only one forested bank, both the magnitude and the absolute volume of
LWD recruited were substantially decreased (table 1). Three of the simulated
patterns (ORND, ODEF, and OPRL) yielded about 50 percent less debris than their
well-forested counterparts. Rather than uniformly decreasing by half the LWD
recruitment totals, the biased patterns tending towards the channel (OTWRD and
OQRT) experienced a decrease of only 15 to 20 percent, suggesting that snags falling
along the major axis are the most important component of riparian LWD recruitment.
Conclusions
This research showed that bank afforestation and snag angle-of-fall significantly
influenced riparian LWD recruitment. Unfortunately, very few simulation studies
have accounted for the impact of streamside forest coverage and snag fall when
predicting long-term woody debris dynamics. With both banks forested, a greater
volume of LWD was delivered over the simulation period, while snag failure patterns
biased towards the stream outproduced random or other patterns not favoring the
channel. Snag angle-of-fall became critical when only one side was forested and a
strong unidirectional control (e.g., prevailing winds) was present.
Acknowledgments
We would like to recognize the following people: Dave Roberts (Utah State
University [USU]); Mark Novak (Bridger-Teton National Forest [BTNF]), Kurt
Nelson (BTNF), Bob Hildebrand (USU), John Shaw (USU), Andy Dolloff, Marty
Spetich, and Bernie Parresol (all of the Southern Research Station, USDA Forest
Service), and several anonymous reviewers. The Fish Ecology Unit of the Forest
Service and the USU Departments of Fisheries and Wildlife and Forest Resources
and the Ecology Center provided financial and logistical support.
68
USDA Forest Service Gen. Tech. Rep. PSW-GTR-181. 2002.
Riparian Large Woody Debris Recruitment—Bragg and Kershner
References
Alexander, R. R.; Buell, J. H. 1955. Determining the direction of destructive winds in a
Rocky Mountain timber stand. Journal of Forestry 53: 19-23.
Berg, D. R. 1995. Riparian silvicultural system design and assessment in the Pacific
Northwest Cascade Mountains, USA. Ecological Applications 5: 87-96.
Bisson, P. A.; Bilby, R. E.; Bryant, M. D.; Dolloff, C. A.; Grette, G. B.; House, R. A.;
Murphy, M. L.; Koski, K. V.; Sedell, J. R. 1987. Large woody debris in forested
streams in the Pacific Northwest: past, present, and future. In: Salo, E. O.; Cundy,
T. W., editors. Streamside management: forestry and fishery interactions. Seattle, WA:
University of Washington Press; 143-190.
Bragg, D. C. 2000. Simulating catastrophic and individualistic large woody debris
recruitment for a small riparian system. Ecology 81: 1383-1394.
Bragg, D. C.; Kershner, J. L.; Roberts, D. W. 2000. Modeling large woody debris in
riparian mixed conifer forests and small streams of the central Rocky Mountains.
Gen. Tech. Rep. RMRS-55. Ogden, UT: Rocky Mountain Research Station, Forest
Service, U.S. Department of Agriculture; 36 p.
Dolloff, C. A. 1994. Large woody debris—the common denominator for integrated
environmental management of forest streams. In: Cairns, J.; Crawford, T. V.;
Salwasser, H., editors. Implementing integrated environmental management.
Blacksburg, VA: Virginia Polytechnic Institute and State University; 93-108.
Grizzel, J. D.; Wolff, N. 1998. Occurrence of windthrow in forest buffer strips and its
effect on small streams in northwest Washington. Northwest Science 72: 214-223.
Hairston-Strang, A. B.; Adams, P. W. 1998. Potential large woody debris sources in
riparian buffers after harvesting in Oregon, USA. Forest Ecology and Management
112: 67-77.
Kershner, J. L. 1997. Setting riparian/aquatic restoration objectives within a watershed
context. Restoration Ecology 5: 15-24.
Maser, C.; Trappe, J. M. 1984. The seen and unseen world of the fallen tree. Gen. Tech.
Rep. PNW-164. Portland, OR: Pacific Northwest Forest Experiment Station, Forest
Service, U.S. Department of Agriculture; 56 p.
Murphy, M. L.; Koski, K. V. 1989. Input and deletion of woody debris in Alaska streams
and implications for streamside management. North American Journal of Fisheries
Management 9: 427-436.
Rainville, R. P.; Rainville, S. C.; Lider, E. L. 1985. Riparian silvicultural strategies for fish
habitat emphasis. Washington, D.C., Proceedings, 1985 Society of American Foresters
annual convention; 186-196.
Robison, E. G.; Beschta, R. L. 1990. Identifying trees in riparian areas that can provide
coarse woody debris to streams. Forest Science 36: 790-801.
Schmid, J. M.; Mata, S. A; McCambridge, W. F. 1985. Natural falling of beetle-killed
ponderosa pine. Res. Note RM-131. Fort Collins, CO: Rocky Mountain Forest and
Range Experiment Station, Forest Service, U.S. Department of Agriculture.
Van Sickle, J.; Gregory, S. V. 1990. Modeling inputs of large woody debris to streams
from falling trees. Canadian Journal of Forest Research 20: 1593-1601.
Veblen, T. T. 1986. Treefalls and the coexistence of conifers in subalpine forests of the
central Rockies. Ecology 67: 644-649.
USDA Forest Service Gen. Tech. Rep. PSW-GTR-181. 2002.
69
Riparian Large Woody Debris Recruitment—Bragg and Kershner
Wykoff, W. R.; Crookston, N. L.; Stage, A. R. 1982. User’s guide to the stand prognosis
model. Gen. Tech. Rep. INT-133. Ogden, UT: Intermountain Forest and Range
Experiment Station, Forest Service, U.S. Department of Agriculture; 112 p.
Zar, J. H. 1984. Biostatistical analysis. 2nd ed. Engelwood Cliffs, NJ: Prentice-Hall; 718 p.
70
USDA Forest Service Gen. Tech. Rep. PSW-GTR-181. 2002.